Nested sampling with normalizing flows for gravitational-wave inference

We present a novel method for sampling iso-likelihood contours in nested using type of machine learning algorithm known as normalising flows and incorporate it into our sampler nessai. Nessai is designed problems where computing the likelihood computationally expensive therefore cost training flow offset by overall reduction number evaluations. validate on 128 simulated gravitational wave signals from compact binary coalescence show that produces unbiased estimates system parameters. Subsequently, we compare results to those obtained with dynesty find good agreement between computed log-evidences whilst requiring 2.07 times fewer also highlight how evaluation can be parallelised nessai without any modifications algorithm. Finally, outline diagnostics included these used tune sampler's settings.